
use Koliev&Lebovic.figure_3.dta

/*
x axis
1    = model 1 results (H1)
2-3  = model 2 results (H2)
4-6  = supplemental results

c (1-6) number used to order the above results from left to right along the x-axis

ap2  = audit prediction-2 (prompt submission)
ap1  = audit prediction-1 (late submission)
ap0  = audit prediction-0 (non-submission)
np2  = audit prediction-2 (prompt submission)
np1  = audit prediction-1 (late submission)
np0  = audit prediction-0 (non-submission)

for confidence intervals

ap2l/ap2h = low/high range for ap2
ap1l/ap1h = low/high range for ap1
ap0l/ap0h = low/high range for ap0
np2l/np2h = low/high range for np2
np1l/np1h = low/high range for np1
np0l/np0h = low/high range for np0
*/

twoway (scatter np2 c) (scatter ap2 c)(rcap np2l np2h c) (rcap ap2l ap2h c)
twoway (scatter np1 c) (scatter ap1 c)(rcap np1l np1h c) (rcap ap1l ap1h c)
twoway (scatter np0 c) (scatter ap0 c)(rcap np0l np0h c) (rcap ap0l ap0h c) 

clear
use Koliev&Lebovic.figure_4.dta

/*
x axis
1    = model 1 results (H1)
2-3  = model 2 results (H2)
4-6  = supplemental results

c (1-6) number used to order the above results from left to right along the x-axis

pp2  = punishment prediction-2 (prompt submission)
np2  = punishment prediction-2 (prompt submission)

for confidence intervals

pp2l/pp2h = low/high range for pp2
np2l/np2h = low/high range for np2
*/


twoway (scatter np2 c) (scatter pp2 c)(rcap np2l np2h c) (rcap pp2l pp2h c)
